| *** sanjayu_ has quit IRC | 00:02 | |
| *** saneax has joined #zuul | 00:06 | |
| *** openstack has joined #zuul | 13:14 | |
| *** ChanServ sets mode: +o openstack | 13:14 | |
| AJaeger | mnaser: AFAIU required-projects checks the repo out. IF the repo is a trusted repo, it's master, not the depends-on change... | 13:15 |
|---|---|---|
| mnaser | AJaeger: ah okay gotcha, so i cant do anything then in that case :) | 13:20 |
| fungi | mnaser: and to be clear, that's a security feature. trusted projects contain security-sensitive data and controls which could be compromised if zuul allowed speculative execution of configuration from a trusted project in an untrusted context (e.g. a pre-review "check" pipeline) | 13:29 |
| jangutter | I believe a while ago I saw something that checked the commit header for things like 'DNM'/'WIP'/'POC' and specifically skipped jobs or reacted differently to the review. Sorry for being vague, but does this trigger anyone's memory? | 13:29 |
| jangutter | Specifically, the use case is to only run linters on reviews with those strings in the header on check, and skip other jobs. | 13:31 |
| fungi | doesn't sound familiar to me. it's also definitely not something i'd want for my workflow since i often push do-not-merge changes specifically to exercise particular jobs | 13:32 |
| jangutter | yeah, I wonder if I'm misremembering something like DNM was not allowed to pass gate... but it could totally be a flight of imagination. | 13:36 |
| *** pcaruana has quit IRC | 13:46 | |
| *** pcaruana has joined #zuul | 13:47 | |
| mnaser | fungi: does that mean using secrets in untrusted jobs is a security issue ? | 13:48 |
| mnaser | Because if so I think I know of a project that is pushing consumable artifacts for an OpenStack project that has in repo secrets... | 13:49 |
| pabelanger | mnaser: no, you can have secrets in untrusted. Just means project needs to be careful about which other projects are allowed to inherit jobs, and ensure nobody merges echo $secret job. Check would be safe, but gate would expose them | 13:51 |
| *** sgw has joined #zuul | 13:57 | |
| corvus | pabelanger: that's not entirely true; secrets are tightly bound to the playbooks in the job they are defined, so child jobs can't access the secrets of parent jobs. | 14:02 |
| corvus | mhu: i don't think we have that yet, but i agree it would be nice. i think atm you would need to crawl all the project definitions using the api | 14:03 |
| pabelanger | corvus: ah right. I was most thinking, if the parent job wrote the secret to disk some place | 14:03 |
| corvus | yes, that requires care and thought | 14:05 |
| fungi | mnaser: the difference is that secrets-using playbooks in trusted repos can be run (non-speculatively) or similarly inherited by jobs in post-review pipelines for untrusted repos. secrets-using playbooks in untrusted repos can't be used by jobs in other untrusted repos (modulo the recently-merged ability for config projects to expressly override the allowed-projects list) | 14:12 |
| fungi | https://zuul-ci.org/docs/zuul/user/config.html#attr-job.allowed-projects | 14:13 |
| mhu | corvus: thanks! | 14:13 |
| *** jeliu_ has joined #zuul | 14:29 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add .gitreview https://review.opendev.org/667942 | 14:38 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add .gitreview and .zuul.yaml https://review.opendev.org/667942 | 14:39 |
| openstackgerrit | James E. Blair proposed zuul/zuul-operator master: Add .gitreview and .zuul.yaml https://review.opendev.org/667942 | 14:41 |
| corvus | i'm self-approving that ^ | 14:42 |
| openstackgerrit | Merged zuul/zuul-operator master: Add .gitreview and .zuul.yaml https://review.opendev.org/667942 | 14:42 |
| fungi | wfm | 14:45 |
| *** bhavikdbavishi has joined #zuul | 14:57 | |
| *** bhavikdbavishi has quit IRC | 15:02 | |
| *** bhavikdbavishi has joined #zuul | 15:10 | |
| *** hashar has joined #zuul | 15:13 | |
| *** bhavikdbavishi has quit IRC | 15:14 | |
| *** pcaruana has quit IRC | 15:14 | |
| *** bhavikdbavishi has joined #zuul | 15:15 | |
| *** gtema_ has joined #zuul | 15:22 | |
| *** portdirect has quit IRC | 15:30 | |
| *** portdirect has joined #zuul | 15:31 | |
| *** chandankumar is now known as raukadah | 15:34 | |
| *** mattw4 has joined #zuul | 15:43 | |
| fbo | hi, is there a way to force zuul ansible_python_interpreter to python3 ? | 15:53 |
| fbo | we have tried to set https://zuul-ci.org/docs/nodepool/configuration.html#attr-diskimages.python-path in nodepool w/o luck | 15:54 |
| corvus | fbo: some changes just went into nodepool to be able to set that on a per-image basis, that's probably the best way to .... | 15:54 |
| corvus | oh :( | 15:54 |
| corvus | fbo: you'll need to be running the newest releases of zuul and nodepool for that (or git master) | 15:55 |
| corvus | just released last week | 15:55 |
| fbo | corvus: ah :) ok that's the reason then. Thanks | 15:55 |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Switch functional testing to a devstack consumer job https://review.opendev.org/665023 | 15:58 |
| *** sgw has quit IRC | 16:01 | |
| *** jangutter has quit IRC | 16:02 | |
| *** jangutter has joined #zuul | 16:02 | |
| *** sshnaidm|ruck is now known as sshnaidm|off | 16:06 | |
| *** pwhalen has quit IRC | 16:09 | |
| *** igordc has joined #zuul | 16:11 | |
| *** sgw has joined #zuul | 16:20 | |
| *** jpena|mtg is now known as jpena|off | 16:21 | |
| *** mattw4 has quit IRC | 16:26 | |
| *** panda has quit IRC | 16:32 | |
| *** hashar has quit IRC | 16:35 | |
| *** panda has joined #zuul | 16:36 | |
| raukadah | Hello | 16:54 |
| raukadah | is it possible to create a symlink in image from nodepool config? | 16:55 |
| raukadah | on openstack provider | 16:55 |
| clarkb | raukadah: yes, the disk image builder elements can create a symlink at image build time | 16:55 |
| tobiash | corvus: is the functional testing change ready for review? | 16:58 |
| clarkb | raukadah: what you'll need to do is createa disk image build element that creates the files and/or symlinks that you want | 16:58 |
| raukadah | clarkb: do I need to use some specific element for that? | 16:58 |
| clarkb | raukadah: you'll likely need to write your own | 16:59 |
| corvus | tobiash: yes! literally right this minute the whole next phase of the stack returned green :) | 16:59 |
| raukadah | I just wanted to create a symlink from /usr/bin/python to /usr/bin/python3 | 16:59 |
| clarkb | raukadah: I would recommend against that | 16:59 |
| corvus | tobiash: (i will rebase the removal change now) | 16:59 |
| clarkb | raukadah: python upstream has stated python2 == python and python3 != python (as far as executable names go) | 16:59 |
| clarkb | raukadah: that will liekly create system confusion if other applications expect python2 at `python` | 17:00 |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Remove devstack plugin functional test jobs https://review.opendev.org/667156 | 17:00 |
| corvus | that's the rebase of the removal | 17:00 |
| raukadah | clarkb: what if all the application assumes python3 == python then | 17:01 |
| clarkb | raukadah: then those applications are buggy | 17:01 |
| clarkb | (unless you are on arch) | 17:01 |
| clarkb | my biggest concern is that if you are on a platform (basically anything but arch) where python == python2 any system applications that expect python to be python2 will break | 17:02 |
| clarkb | and so would be preferable to tell python3 applications to use `python3` | 17:02 |
| clarkb | another potential option is to use a virtualenv | 17:02 |
| clarkb | (since that will isolate the system python concerns) | 17:02 |
| clarkb | but in any case you'll need an element | 17:03 |
| clarkb | there is no generic make a symlink element | 17:03 |
| corvus | oh, my, the tentacles extend to openstacksdk too | 17:03 |
| raukadah | clarkb: ok, it is the same case which fbo was discussing earlier | 17:03 |
| clarkb | raukadah: ok so for that you can set that as an attribute of the nodepool config, but it tells ansible to use `python3` not a symlink from python to python3 | 17:04 |
| clarkb | let me find a link | 17:04 |
| *** jeliu_ has quit IRC | 17:05 | |
| clarkb | raukadah: https://zuul-ci.org/docs/nodepool/configuration.html#attr-diskimages.python-path set that to /usr/bin/python3 | 17:05 |
| clarkb | (this requires an up to date nodepool and zuul install, most recent releases for both) | 17:06 |
| *** hwangbo has quit IRC | 17:06 | |
| raukadah | clarkb: the nodepool and zuul is not update in sf | 17:07 |
| raukadah | clarkb: https://softwarefactory-project.io/r/#/c/15834/ I tried that so got suggested to create a symlink | 17:07 |
| pabelanger | you can set ansible_python_interpreter in your job, if you need to change it too | 17:10 |
| pabelanger | that doesn't need a new release of nodepool | 17:10 |
| clarkb | pabelanger: I think zuul overrides it though | 17:10 |
| corvus | nope | 17:11 |
| pabelanger | clarkb: nope | 17:11 |
| corvus | we use it in opendev | 17:11 |
| clarkb | oh so the hardcoded value in zuul prior to the recent release can be overridden in the job directly? | 17:11 |
| clarkb | Oh the hardcoded value is what the executor runs but the rmeote node can be set got it | 17:11 |
| clarkb | or something like that, the remote python is determined at runtime, not prior to starting ansible on the executor | 17:12 |
| *** hwangbo has joined #zuul | 17:14 | |
| corvus | clarkb: i think the locality of setting it as a host-var overrode the hardcoded global setting | 17:14 |
| raukadah | clarkb: pabelanger https://review.rdoproject.org/r/#/c/21270/4/zuul.d/standalone-jobs.yaml@26 | 17:15 |
| raukadah | i have tried this | 17:15 |
| pabelanger | clarkb: yah, it gets a little tricky, with more then 1 nodeset using the job. I kinda wish you could set a nodeset.vars, but with new python-path in nodepool, that isn't needed | 17:16 |
| pabelanger | raukadah: is there logs from the job run? | 17:17 |
| clarkb | out of curiousity what is the problem with using python2 there? Does ansible not run under rhel8's python2? | 17:17 |
| clarkb | or maybe we are trying to avoid installing python2 on rhel8? | 17:18 |
| pabelanger | there is no python2 on rhel8 | 17:18 |
| clarkb | oh I thought there was a python2 | 17:18 |
| clarkb | but primary was 3 | 17:18 |
| pabelanger | no, last I heard it was all removed | 17:18 |
| clarkb | the internet tells me that you can install python2 on rhel8 | 17:19 |
| pabelanger | however, I haven't tried rhel8 GA | 17:19 |
| clarkb | it isn't there by default but it is available as a package | 17:19 |
| pabelanger | ah | 17:19 |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Remove devstack plugin functional test jobs https://review.opendev.org/667156 | 17:19 |
| raukadah | trying to capture the stream out, the job is queued | 17:19 |
| pabelanger | clarkb: my guess, is product reasons not to install python2 | 17:19 |
| pabelanger | but, that is a wild guess | 17:19 |
| fungi | of course, having your job install python2 so that ansible can use it is a bit of a catch-22, so would probably need to install it in the image instead | 17:20 |
| clarkb | fungi: yes would have to be part of the image build | 17:20 |
| pabelanger | there is some good news, with ansible 2.8, they've make it a little smarter to look for python the remote node. | 17:21 |
| pabelanger | There is a new 'auto' setting you can setup for ansible_python_interpreter that will use any python found | 17:21 |
| pabelanger | it searches multiple paths for it | 17:21 |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Remove devstack plugin functional test jobs https://review.opendev.org/667156 | 17:21 |
| corvus | we have oodles of ways of selecting the correct python for ansible in zuul, so noone should need to update base images | 17:24 |
| tobiash | corvus: I left a question on 665023 | 17:24 |
| clarkb | corvus: ya the question was more why is a symlink preferred to actually installing python2 | 17:25 |
| clarkb | sounds like it should be possible to address without changing the image though so yay | 17:25 |
| corvus | tobiash: replied, thanks | 17:26 |
| tobiash | ++ | 17:27 |
| openstackgerrit | Dirk Mueller proposed zuul/nodepool master: Add Python 3.7 testing https://review.opendev.org/667720 | 17:29 |
| *** rlandy is now known as rlandy|brb | 17:29 | |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Reparent nodepool-functional-openstack-src https://review.opendev.org/667995 | 17:29 |
| openstackgerrit | James E. Blair proposed zuul/nodepool master: Reparent nodepool-functional-openstack-src https://review.opendev.org/667995 | 17:37 |
| *** mattw4 has joined #zuul | 17:40 | |
| corvus | tobiash: fyi see -1 on 667720 | 17:41 |
| tobiash | Oh, that's right | 17:43 |
| openstackgerrit | Dirk Mueller proposed zuul/nodepool master: Add Python 3.7 testing https://review.opendev.org/667720 | 17:43 |
| *** rlandy|brb is now known as rlandy | 17:45 | |
| *** gtema_ has quit IRC | 17:47 | |
| *** bhavikdbavishi has quit IRC | 17:48 | |
| *** mattw4 has quit IRC | 17:50 | |
| openstackgerrit | Dirk Mueller proposed zuul/zuul master: Add Python 3.7 testing https://review.opendev.org/668006 | 17:52 |
| *** electrofelix has quit IRC | 17:53 | |
| *** pcaruana has joined #zuul | 17:54 | |
| *** mattw4 has joined #zuul | 18:12 | |
| *** jeliu_ has joined #zuul | 18:15 | |
| *** mattw4 has quit IRC | 18:17 | |
| *** mattw4 has joined #zuul | 18:24 | |
| *** mattw4 has quit IRC | 18:30 | |
| *** themroc has quit IRC | 18:44 | |
| SpamapS | so | 18:47 |
| SpamapS | we're broken right now | 18:48 |
| SpamapS | install-nodejs | 18:48 |
| SpamapS | I've never heard of this 'command' command, and neither have my servers. :-/ | 18:50 |
| fungi | i lack context | 18:50 |
| openstackgerrit | Clint 'SpamapS' Byrum proposed zuul/zuul-jobs master: Revert "install-nodejs: add support for RPM-based OSes" https://review.opendev.org/668021 | 18:50 |
| fungi | now i have context! | 18:50 |
| fungi | this is the first i've heard of `command` but it seems to be a shell builtin? | 18:53 |
| fungi | command [-pVv] command [arg ...] Run command with args suppressing the normal shell function lookup. Only builtin commands or commands found in the PATH are executed. [...] | 18:54 |
| clarkb | yes looks like bash builtin | 18:54 |
| fungi | from bash(1) | 18:54 |
| fungi | does not seem to be posix bourne | 18:55 |
| fungi | maybe copied from zsh? | 18:55 |
| fungi | at any rate, we should probably not assume an ansible shell task is running in a shell which has that builtin | 18:56 |
| clarkb | I'm probably blind but where is `command` in that revert? | 18:56 |
| fungi | roles/install-nodejs/tasks/main.yaml | 18:56 |
| fungi | command: command -v node | 18:57 |
| clarkb | bah gerrit search only searches the new side of the diff | 18:57 |
| fungi | If the -V or -v option is supplied, the exit status is 0 if command was found, and 1 if not. | 18:57 |
| clarkb | type also works but that is anohter builtin. In any case should use a shell ratherthan command task I agree | 18:57 |
| fungi | apparently a bashism for which(1) | 18:57 |
| fungi | the manpage for which(1) indicates this is its default behavior | 18:58 |
| fungi | EXIT STATUS: 0 if all specified commands are found and executable; 1 if one or more specified commands is nonexistent or not executable | 18:59 |
| clarkb | I think the struggle with which is it isnt always installed | 18:59 |
| fungi | ahh | 18:59 |
| clarkb | so the tradeoff here is expecting bash to exist or expecting which to exist | 18:59 |
| clarkb | personally I expect both to exist but I know that is not true of all installations | 18:59 |
| corvus | SpamapS: +2 with a comment that we should add jobs to test that role when we next take a crack at it. but we should nail down the particulars of your environment to make sure we have it covered. | 19:00 |
| fungi | does ansible not have a feature to test for the existence of an executable in the path? | 19:00 |
| clarkb | I've approved the revert | 19:00 |
| clarkb | fungi: I am not aware of one | 19:00 |
| corvus | sounds like a system with non-bash as the default probably would have caught that | 19:00 |
| clarkb | corvus: ya I think ubuntu would've caught it | 19:01 |
| * clarkb checks if dash has command | 19:01 | |
| fungi | double-approved i guess | 19:01 |
| clarkb | oh wow dash has command too | 19:01 |
| corvus | it's possible that a solution which would satisfy everything would be to specify those ansible shell tasks run with bash | 19:01 |
| clarkb | corvus: in this case it is a command task which is extra problematic | 19:02 |
| fungi | ahh, yeah, command doesn't run in the context of a shell at all, right? | 19:02 |
| corvus | er, then if it does exist, what exactly does ansible run? | 19:03 |
| corvus | (i thought a shell builtin would have to be run in a shell) | 19:03 |
| clarkb | corvus: yes it should be in a shell task | 19:03 |
| clarkb | but then we'd also have to ensure that the shell run (which is sh by default which varies on all platforms) has `command` or similar | 19:04 |
| clarkb | my suse sh is bash. ubuntu sh is dash. I know some docker images have bourne shell as sh and so on | 19:04 |
| corvus | so does "command: command" even work? | 19:04 |
| corvus | with "shell:" you can specify which shell to use | 19:05 |
| *** gtema_ has joined #zuul | 19:05 | |
| clarkb | if ansible runs command in a shell then it will work if that shell has command, I don't know if ansible runs command tasks in a shell or not | 19:05 |
| corvus | ({shell: 'command foo', args: {executable: '/bin/bash}}) | 19:05 |
| corvus | clarkb: interesting question; i assumed since it didn't allow a shell argument that it did not use a shell, and was like an exec.... | 19:06 |
| corvus | fatal: [localhost]: FAILED! => {"changed": false, "cmd": "command node", "msg": "[Errno 2] No such file or directory: 'command': 'command'", "rc": 2} | 19:07 |
| clarkb | so ya I don't think that ever worked then | 19:07 |
| fungi | agreed | 19:07 |
| corvus | moving the zuul-jobs test jobs into zuul-jobs is my next move-to-zuul-tenant task, so i'll try to set up more docs/examples around that, and we can start to improve testing there | 19:09 |
| openstackgerrit | Merged zuul/zuul-jobs master: Revert "install-nodejs: add support for RPM-based OSes" https://review.opendev.org/668021 | 19:19 |
| corvus | SpamapS: ^ | 19:19 |
| openstackgerrit | Jeff Liu proposed zuul/zuul-operator master: Add playbooks folder and debug zuul job https://review.opendev.org/668029 | 19:20 |
| corvus | tobiash, SpamapS, mordred, tristanC: jeliu_ is an intern at red hat who is starting to help us with work on the operator... for now, we're working on setting up some basic testing infrastructure, then as the spec comes together, he'll be working on fleshing out the implementation | 19:20 |
| *** gtema_ has quit IRC | 19:21 | |
| *** gtema_ has joined #zuul | 19:22 | |
| fungi | that's exciting. welcome jeliu_! | 19:26 |
| tobiash | Welcome jeliu_ ! | 19:28 |
| jeliu_ | Thanks! Excited to be here! | 19:29 |
| SpamapS | corvus:thanks so much for reverting so quickly! | 19:33 |
| SpamapS | Probably should add a local gated mirror to make these less urgent. | 19:33 |
| SpamapS | Is there any reason to use the `command` command in the `command` module in Ansible? | 19:34 |
| tobiash | So many commands | 19:35 |
| SpamapS | I command you to use command. | 19:36 |
| clarkb | SpamapS: no I think it will only work when run in a shell (and only if that shell has that builtin so not bourne or csh (probably)) | 19:37 |
| clarkb | jeliu_: hello | 19:37 |
| clarkb | and welcome | 19:37 |
| SpamapS | So the real problem is that the code path isn't tested. | 19:37 |
| fungi | SpamapS: yeah, we don't currently exercise every role in zuul-jobs on changes to zuul-jobs | 19:38 |
| jeliu_ | clarkb Thanks! | 19:39 |
| fungi | SpamapS: corvus has some ideas on how we can do functional testing of roles and playbooks (i think as a follow on to the current integration testing?) | 19:40 |
| fungi | and was looking for volunteers to work on implementation | 19:40 |
| SpamapS | I'd be quite happy to be a 3rd party test. | 19:40 |
| SpamapS | one of the other engs here noticed another problem with the change | 19:42 |
| fungi | awesome | 19:42 |
| SpamapS | the original installed apt-transport-https, but the new one's Debian tasks dont. | 19:42 |
| fungi | which is okay for newer debian releases, but probably not <buster? | 19:43 |
| clarkb | or <= xenial | 19:43 |
| fungi | (i can't remember when https went into the apt package) | 19:43 |
| SpamapS | + install_nodejs_node_installed.rc == 1 or install_nodejs_npm_installed == 1 | 19:44 |
| SpamapS | that's also a bug | 19:44 |
| fungi | yeah, https://packages.debian.org/apt-transport-https confirms it was an actual package through stretch, and only becomes a transitional package as of buster | 19:44 |
| SpamapS | it should be != 0, not == 1 | 19:45 |
| SpamapS | and the second one needs .rc | 19:45 |
| SpamapS | That led to the conditional result failing | 19:46 |
| SpamapS | jeliu_:hey howdy. Sorry for pointing out bugs publicly. I'm sure this was a challenging task. We'll pile on to make sure the next round goes well. :) | 19:47 |
| SpamapS | Also I just want to remind everyone that zuul-jobs is not tagged, doesn't do release controls, and as a result, can really screw things up for users who are consuming it directly. Careful reviews are definitely needed. | 19:47 |
| fungi | yep. can still leave comments about all these points on https://review.opendev.org/631049 | 19:47 |
| fungi | even though it's merged (and reverted) | 19:47 |
| SpamapS | (I am going to resurrect my old task to move to a gated mirror for our deploy) | 19:48 |
| fungi | SpamapS: jeliu_ is working on the zuul-operator for kubernetes. the patch which got reverted was mhu's | 19:48 |
| *** hwangbo has quit IRC | 19:49 | |
| *** hwangbo has joined #zuul | 19:50 | |
| SpamapS | oh | 19:51 |
| SpamapS | derp | 19:51 |
| hwangbo | how much work would need to be done to support dependency cycles? I saw that it was discussed roughly in this thread http://lists.zuul-ci.org/pipermail/zuul-discuss/2018-November/000637.html | 19:57 |
| clarkb | hwangbo: I believe it requires adding direct merge support to zuul | 19:57 |
| clarkb | because hitting the merge button in gerrit/github would sequence changes/PRs | 19:58 |
| clarkb | and could fail halfway through | 19:59 |
| *** hwangbo has quit IRC | 20:04 | |
| *** hwangbo has joined #zuul | 20:04 | |
| *** hwangbo has quit IRC | 20:05 | |
| *** hwangbo has joined #zuul | 20:05 | |
| fungi | yes, the risk is that you decide to merge changes a,b,c together and then you call gerrit to submit a and it works and then you call gerrit to submit b and it's rejected, and you're left with a merged but not b or c | 20:05 |
| fungi | so zuul would need to push the merge commits into gerrit instead to avoid that | 20:06 |
| *** jeliu_ has quit IRC | 20:09 | |
| *** hwangbo has quit IRC | 20:23 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Split job definitions into multiple files https://review.opendev.org/668040 | 20:36 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add a simple test of the install-nodejs role https://review.opendev.org/668043 | 20:43 |
| corvus | i think i'd like to add a feature which allows an admin to specify that zuul should load in-repo config from a non-standard location for a particular project. that way, we can put a project stanza in zuul-jobs, but put it in a file like "zuul-jobs-testing.d/project.yaml", and tell the opendev zuul to additionally load files in "zuul-jobs-testing.d" just for the zuul-jobs repo (and just in the zuul tenant). | 20:46 |
| corvus | that way no other users of zuul-jobs see that file, but we can still do in-repo config of the project stanza for zuul-jobs. | 20:46 |
| corvus | add to that the change i want to do to have a files-matcher which automatically matches if the current job config is changed, and i think the testing story for zuul-jobs will get even better | 20:46 |
| corvus | i bet we could add https://review.opendev.org/668043 to 75% of our roles without writing any additional playbook code and get Pretty Good Coverage (TM) | 20:52 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: DNM: install-nodejs: add support for RPM-based OSes https://review.opendev.org/668048 | 20:53 |
| corvus | in fact... let's see if it would have caught that ^ | 20:54 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Move upload-git-mirror test job in-repo https://review.opendev.org/668050 | 21:01 |
| corvus | fungi, clarkb, SpamapS: check it out! http://logs.openstack.org/48/668048/1/check/zuul-jobs-test-install-nodejs/dd2dab4/ara-report/result/ce8c4427-7c3f-463a-9e26-7276869bf8d3/ | 21:03 |
| *** gtema_ has quit IRC | 21:04 | |
| corvus | i'm waiting for the git-mirror job results to arrive, then i'll push up a final version of that for review | 21:05 |
| *** pcaruana has quit IRC | 21:05 | |
| *** sgw has quit IRC | 21:12 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add a simple test of the install-nodejs role https://review.opendev.org/668043 | 21:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Move upload-git-mirror test job in-repo https://review.opendev.org/668050 | 21:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add zuul-jobs-tox-linters https://review.opendev.org/668052 | 21:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Move upload-git-mirror test job in-repo https://review.opendev.org/668050 | 21:16 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add zuul-jobs-tox-linters https://review.opendev.org/668052 | 21:16 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Move upload-git-mirror test job in-repo https://review.opendev.org/668050 | 21:27 |
| openstackgerrit | Merged zuul/nodepool master: Switch functional testing to a devstack consumer job https://review.opendev.org/665023 | 21:39 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add base integration roles https://review.opendev.org/668061 | 21:44 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add base integration roles https://review.opendev.org/668061 | 22:00 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Move upload-git-mirror test job in-repo https://review.opendev.org/668050 | 22:03 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: Add base integration roles https://review.opendev.org/668061 | 22:03 |
| *** mattw4 has joined #zuul | 22:23 | |
| *** mattw4 has quit IRC | 22:28 | |
| *** mattw4 has joined #zuul | 22:28 | |
| *** igordc has quit IRC | 22:36 | |
| *** mattw4 has quit IRC | 22:37 | |
| *** rlandy has quit IRC | 22:57 | |
| SpamapS | corvus:zomg I love the idea of setting the paths to look for config in the tenant config stanza for the project. That kind of works in the same way I wanted namespacing to work. | 22:58 |
| SpamapS | corvus:also nice job on the test! | 22:59 |
| *** tosky has quit IRC | 23:00 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!